Leveraging EfficientNet and Contrastive Learning for Accurate Global-scale Location Estimation

TL;DR

This paper introduces a unified approach combining EfficientNet-based classification and contrastive learning for image geolocation, achieving state-of-the-art accuracy at fine spatial scales.

Contribution

It proposes a novel mixed classification-retrieval scheme that leverages EfficientNet and contrastive learning for improved global image geolocation.

Findings

Achieves 15.0% accuracy at 1km on Im2GPS3k

Demonstrates superior performance on four public datasets

Sets new state-of-the-art in fine granularity geolocation

Abstract

In this paper, we address the problem of global-scale image geolocation, proposing a mixed classification-retrieval scheme. Unlike other methods that strictly tackle the problem as a classification or retrieval task, we combine the two practices in a unified solution leveraging the advantages of each approach with two different modules. The first leverages the EfficientNet architecture to assign images to a specific geographic cell in a robust way. The second introduces a new residual architecture that is trained with contrastive learning to map input images to an embedding space that minimizes the pairwise geodesic distance of same-location images. For the final location estimation, the two modules are combined with a search-within-cell scheme, where the locations of most similar images from the predicted geographic cell are aggregated based on a spatial clustering scheme. Our approach demonstrates very competitive performance on four public datasets, achieving new state-of-the-art performance in fine granularity scales, i.e., 15.0% at 1km range on Im2GPS3k.